BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
pcapazzi
Pyrite | Level 9

I have a campaign that pulls a list of customers that is in a table that with several different lists in it for the subject (prospects, dnc, preferences, etc.). When I get to the split node to create the segments the node finds the subjects in other lists. Simply limiting the subject to one of the segments by priority will place the subject in the wrong segment after a few occurrences. I solutioned this by using a copy of the select node just before the split and after the node from the susequent cell (that was before the split). 

 

It'd be ideal if the split node had a refine output option similar to Export and the Communication node's Export function. 

 

Open to suggestions as well. 

 

1 ACCEPTED SOLUTION

Accepted Solutions
SteveMarshall
SAS Employee

pcapazzi,

You are correct that 1 or more selects immediately upstream from a Split node can be used to filter the rows available for a split based on Data Item.  Your approach is the proper way of controlling/filtering rows from the Data Item's table that is many-1 with the subject during split processing.    Your suggestion to enable a Refine capability similar to Export for Split on Data Item would also provide this capability.   

 

I do have a concern regarding your setup.  You say that "the List table is all you're using in the Information Map for this campaign".  Hopefully you have a Subject table that has only 1 row per Subject Key in addition to the List Table in your information map.

 

There are other times where replicating a select node can be beneficial.  Suppose your database tables are distributed / partitioned based on a common field (say Sub_Business).  By replicating the select node (Sub_Business=123) after cells and before subsequent Selects of Splits, you can ensure that the database understands that it only needs to process on partitions associated with that value.

 

View solution in original post

4 REPLIES 4
shill
SAS Employee

I'm not sure I completely understand the scenario. At first glance it does sound like a job for the prioritize node, which would mean you'd want multiple select nodes (select where list type = prospect, select where list type = dnc, etc.) that all connect to a prioritize node which results in a the same number of output cells but with mutually exclusive subject IDs.

 

I'm guessing there's something I'm missing in the simple approach, though. Can you provide some more context if this isn't a viable solution for you?


Register today and join us virtually on June 16!
sasglobalforum.com | #SASGF

View now: on-demand content for SAS users

pcapazzi
Pyrite | Level 9

Hi shill,

I probably could've explained it better. I hope this helps. 

 

Say you have a table with subjects and various lists. You as subject 12345 exist in the list 1 which is 'campaign 1'. Campaign 1 has 5 touchpoints to it spread out over 5 weeks. You have a date added of last Wednesday. You also exist in the same list table in list 999 which is 'DNC' and you have a date you were entered there as 3 Wednesdays ago. 

 

This list table is all I'm using in the informaiton map for this campaign. When I get to the split node I can only select one field to split by which is the week of the date you were added. If you are 1 week old you go into the 1st touchpoint segment (cell). If you were added 2 weeks ago you would go to the 2nd week segment. After 5 weeks you would not be receiving any more marketing material for the campaign. 

 

Since the split node operates on one field it doesn't have anything to limit it to the list selected from the initial Select node (list 1). It finds you also in list 999 from 3 weeks ago so you pop up as a duplicate subject that lands in the week 3 segment. 

 

Now, reverse the entry points to the lists. Say you were added to the campaign list (list 1) 3 weeks ago and the DNC list 1 week ago. If you you a prioritize node how can you give it an order to keep the subject from the list instead of DNC? If it's only operating on week number it won't know to ignore the week number for any other than list 1. 

 

If a Refine Output as we see in the Export or Communication nodes we could lock it down to the subjects from that list. 

 

Thanks,

SteveMarshall
SAS Employee

pcapazzi,

You are correct that 1 or more selects immediately upstream from a Split node can be used to filter the rows available for a split based on Data Item.  Your approach is the proper way of controlling/filtering rows from the Data Item's table that is many-1 with the subject during split processing.    Your suggestion to enable a Refine capability similar to Export for Split on Data Item would also provide this capability.   

 

I do have a concern regarding your setup.  You say that "the List table is all you're using in the Information Map for this campaign".  Hopefully you have a Subject table that has only 1 row per Subject Key in addition to the List Table in your information map.

 

There are other times where replicating a select node can be beneficial.  Suppose your database tables are distributed / partitioned based on a common field (say Sub_Business).  By replicating the select node (Sub_Business=123) after cells and before subsequent Selects of Splits, you can ensure that the database understands that it only needs to process on partitions associated with that value.

 

pcapazzi
Pyrite | Level 9

Thanks... yes... what I meant was that in the campaign diagram I'm using only the List Table in various select nodes. There is a subject table in the information map with a single row per subject. 

How to improve email deliverability

SAS' Peter Ansbacher shows you how to use the dashboard in SAS Customer Intelligence 360 for better results.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 4 replies
  • 2667 views
  • 0 likes
  • 3 in conversation